5th International Conference on Spoken Language Processing

Sydney, Australia
November 30 - December 4, 1998

SQEL: A Multilingual and Multifunctional Dialogue System

Maria Aretoulaki (1), Stefan Harbeck (2), Florian Gallwitz (2), Elmar Nöth (2), Heinrich Niemann (3), Jozef Ivanecky (4), Ivo Ipsic (5), Nikola Pavesic (5), Vaclav Matousek (6)

(1) FORWISS (Bavarian Research Centre for Knowledge-Based Systems), Germany
(2) University of Erlangen, Germany
(3) University of Erlangen & FORWISS, Germany
(4) Technical University of Kosice, Slovak Republic
(5) University of Ljubljana, Slovenia
(6) University of West Bohemia in Pilsen, Czech Republic

Within the EC SQEL project, the German EVAR spoken dialogue system has been extended to handle four different languages and domains: German, Slovak, and Czech (national train connections), and Slovenian (European flights). The SQEL demonstrator can also access databases on the WWW, enabling users without an internet connection to meet their information needs by just using the phone. When the system starts up, the caller is free to use any of the implemented languages. A multilingual word recognizer implicitly identifies the language, which is then associated with the appropriate domain and database. For the remainder of the dialogue, the corresponding monolingual recognizer is employed. The existence of language-independent parameters (e.g. goal/source location) has meant that porting the system to a new language does not involve an extensive restructuring of the interpretation process within the Dialogue Manager. This is sufficiently flexible to switch between the different domains and languages.

Full Paper

Bibliographic reference.  Aretoulaki, Maria / Harbeck, Stefan / Gallwitz, Florian / Nöth, Elmar / Niemann, Heinrich / Ivanecky, Jozef / Ipsic, Ivo / Pavesic, Nikola / Matousek, Vaclav (1998): "SQEL: a multilingual and multifunctional dialogue system", In ICSLP-1998, paper 0329.